1
0
Fork 0
alistair23-linux/drivers/video/fbdev/omap2/omapfb/omapfb-main.c

2657 lines
59 KiB
C
Raw Normal View History

/*
* linux/drivers/video/omap2/omapfb-main.c
*
* Copyright (C) 2008 Nokia Corporation
* Author: Tomi Valkeinen <tomi.valkeinen@nokia.com>
*
* Some code and ideas taken from drivers/video/omap/ driver
* by Imre Deak.
*
* This program is free software; you can redistribute it and/or modify it
* under the terms of the GNU General Public License version 2 as published by
* the Free Software Foundation.
*
* This program is distributed in the hope that it will be useful, but WITHOUT
* ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
* FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for
* more details.
*
* You should have received a copy of the GNU General Public License along with
* this program. If not, see <http://www.gnu.org/licenses/>.
*/
#include <linux/module.h>
#include <linux/delay.h>
include cleanup: Update gfp.h and slab.h includes to prepare for breaking implicit slab.h inclusion from percpu.h percpu.h is included by sched.h and module.h and thus ends up being included when building most .c files. percpu.h includes slab.h which in turn includes gfp.h making everything defined by the two files universally available and complicating inclusion dependencies. percpu.h -> slab.h dependency is about to be removed. Prepare for this change by updating users of gfp and slab facilities include those headers directly instead of assuming availability. As this conversion needs to touch large number of source files, the following script is used as the basis of conversion. http://userweb.kernel.org/~tj/misc/slabh-sweep.py The script does the followings. * Scan files for gfp and slab usages and update includes such that only the necessary includes are there. ie. if only gfp is used, gfp.h, if slab is used, slab.h. * When the script inserts a new include, it looks at the include blocks and try to put the new include such that its order conforms to its surrounding. It's put in the include block which contains core kernel includes, in the same order that the rest are ordered - alphabetical, Christmas tree, rev-Xmas-tree or at the end if there doesn't seem to be any matching order. * If the script can't find a place to put a new include (mostly because the file doesn't have fitting include block), it prints out an error message indicating which .h file needs to be added to the file. The conversion was done in the following steps. 1. The initial automatic conversion of all .c files updated slightly over 4000 files, deleting around 700 includes and adding ~480 gfp.h and ~3000 slab.h inclusions. The script emitted errors for ~400 files. 2. Each error was manually checked. Some didn't need the inclusion, some needed manual addition while adding it to implementation .h or embedding .c file was more appropriate for others. This step added inclusions to around 150 files. 3. The script was run again and the output was compared to the edits from #2 to make sure no file was left behind. 4. Several build tests were done and a couple of problems were fixed. e.g. lib/decompress_*.c used malloc/free() wrappers around slab APIs requiring slab.h to be added manually. 5. The script was run on all .h files but without automatically editing them as sprinkling gfp.h and slab.h inclusions around .h files could easily lead to inclusion dependency hell. Most gfp.h inclusion directives were ignored as stuff from gfp.h was usually wildly available and often used in preprocessor macros. Each slab.h inclusion directive was examined and added manually as necessary. 6. percpu.h was updated not to include slab.h. 7. Build test were done on the following configurations and failures were fixed. CONFIG_GCOV_KERNEL was turned off for all tests (as my distributed build env didn't work with gcov compiles) and a few more options had to be turned off depending on archs to make things build (like ipr on powerpc/64 which failed due to missing writeq). * x86 and x86_64 UP and SMP allmodconfig and a custom test config. * powerpc and powerpc64 SMP allmodconfig * sparc and sparc64 SMP allmodconfig * ia64 SMP allmodconfig * s390 SMP allmodconfig * alpha SMP allmodconfig * um on x86_64 SMP allmodconfig 8. percpu.h modifications were reverted so that it could be applied as a separate patch and serve as bisection point. Given the fact that I had only a couple of failures from tests on step 6, I'm fairly confident about the coverage of this conversion patch. If there is a breakage, it's likely to be something in one of the arch headers which should be easily discoverable easily on most builds of the specific arch. Signed-off-by: Tejun Heo <tj@kernel.org> Guess-its-ok-by: Christoph Lameter <cl@linux-foundation.org> Cc: Ingo Molnar <mingo@redhat.com> Cc: Lee Schermerhorn <Lee.Schermerhorn@hp.com>
2010-03-24 02:04:11 -06:00
#include <linux/slab.h>
#include <linux/fb.h>
#include <linux/dma-mapping.h>
#include <linux/vmalloc.h>
#include <linux/device.h>
#include <linux/platform_device.h>
#include <linux/omapfb.h>
#include <video/omapfb_dss.h>
#include <video/omapvrfb.h>
#include "omapfb.h"
#define MODULE_NAME "omapfb"
#define OMAPFB_PLANE_XRES_MIN 8
#define OMAPFB_PLANE_YRES_MIN 8
static char *def_mode;
static char *def_vram;
static bool def_vrfb;
static int def_rotate;
static bool def_mirror;
static bool auto_update;
static unsigned int auto_update_freq;
module_param(auto_update, bool, 0);
module_param(auto_update_freq, uint, 0644);
#ifdef DEBUG
bool omapfb_debug;
module_param_named(debug, omapfb_debug, bool, 0644);
static bool omapfb_test_pattern;
module_param_named(test, omapfb_test_pattern, bool, 0644);
#endif
static int omapfb_fb_init(struct omapfb2_device *fbdev, struct fb_info *fbi);
static int omapfb_get_recommended_bpp(struct omapfb2_device *fbdev,
struct omap_dss_device *dssdev);
#ifdef DEBUG
static void draw_pixel(struct fb_info *fbi, int x, int y, unsigned color)
{
struct fb_var_screeninfo *var = &fbi->var;
struct fb_fix_screeninfo *fix = &fbi->fix;
void __iomem *addr = fbi->screen_base;
const unsigned bytespp = var->bits_per_pixel >> 3;
const unsigned line_len = fix->line_length / bytespp;
int r = (color >> 16) & 0xff;
int g = (color >> 8) & 0xff;
int b = (color >> 0) & 0xff;
if (var->bits_per_pixel == 16) {
u16 __iomem *p = (u16 __iomem *)addr;
p += y * line_len + x;
r = r * 32 / 256;
g = g * 64 / 256;
b = b * 32 / 256;
__raw_writew((r << 11) | (g << 5) | (b << 0), p);
} else if (var->bits_per_pixel == 24) {
u8 __iomem *p = (u8 __iomem *)addr;
p += (y * line_len + x) * 3;
__raw_writeb(b, p + 0);
__raw_writeb(g, p + 1);
__raw_writeb(r, p + 2);
} else if (var->bits_per_pixel == 32) {
u32 __iomem *p = (u32 __iomem *)addr;
p += y * line_len + x;
__raw_writel(color, p);
}
}
static void fill_fb(struct fb_info *fbi)
{
struct fb_var_screeninfo *var = &fbi->var;
const short w = var->xres_virtual;
const short h = var->yres_virtual;
void __iomem *addr = fbi->screen_base;
int y, x;
if (!addr)
return;
DBG("fill_fb %dx%d, line_len %d bytes\n", w, h, fbi->fix.line_length);
for (y = 0; y < h; y++) {
for (x = 0; x < w; x++) {
if (x < 20 && y < 20)
draw_pixel(fbi, x, y, 0xffffff);
else if (x < 20 && (y > 20 && y < h - 20))
draw_pixel(fbi, x, y, 0xff);
else if (y < 20 && (x > 20 && x < w - 20))
draw_pixel(fbi, x, y, 0xff00);
else if (x > w - 20 && (y > 20 && y < h - 20))
draw_pixel(fbi, x, y, 0xff0000);
else if (y > h - 20 && (x > 20 && x < w - 20))
draw_pixel(fbi, x, y, 0xffff00);
else if (x == 20 || x == w - 20 ||
y == 20 || y == h - 20)
draw_pixel(fbi, x, y, 0xffffff);
else if (x == y || w - x == h - y)
draw_pixel(fbi, x, y, 0xff00ff);
else if (w - x == y || x == h - y)
draw_pixel(fbi, x, y, 0x00ffff);
else if (x > 20 && y > 20 && x < w - 20 && y < h - 20) {
int t = x * 3 / w;
unsigned r = 0, g = 0, b = 0;
unsigned c;
if (var->bits_per_pixel == 16) {
if (t == 0)
b = (y % 32) * 256 / 32;
else if (t == 1)
g = (y % 64) * 256 / 64;
else if (t == 2)
r = (y % 32) * 256 / 32;
} else {
if (t == 0)
b = (y % 256);
else if (t == 1)
g = (y % 256);
else if (t == 2)
r = (y % 256);
}
c = (r << 16) | (g << 8) | (b << 0);
draw_pixel(fbi, x, y, c);
} else {
draw_pixel(fbi, x, y, 0);
}
}
}
}
#endif
static unsigned omapfb_get_vrfb_offset(const struct omapfb_info *ofbi, int rot)
{
const struct vrfb *vrfb = &ofbi->region->vrfb;
unsigned offset;
switch (rot) {
case FB_ROTATE_UR:
offset = 0;
break;
case FB_ROTATE_CW:
offset = vrfb->yoffset;
break;
case FB_ROTATE_UD:
offset = vrfb->yoffset * OMAP_VRFB_LINE_LEN + vrfb->xoffset;
break;
case FB_ROTATE_CCW:
offset = vrfb->xoffset * OMAP_VRFB_LINE_LEN;
break;
default:
BUG();
return 0;
}
offset *= vrfb->bytespp;
return offset;
}
static u32 omapfb_get_region_rot_paddr(const struct omapfb_info *ofbi, int rot)
{
if (ofbi->rotation_type == OMAP_DSS_ROT_VRFB) {
return ofbi->region->vrfb.paddr[rot]
+ omapfb_get_vrfb_offset(ofbi, rot);
} else {
return ofbi->region->paddr;
}
}
static u32 omapfb_get_region_paddr(const struct omapfb_info *ofbi)
{
if (ofbi->rotation_type == OMAP_DSS_ROT_VRFB)
return ofbi->region->vrfb.paddr[0];
else
return ofbi->region->paddr;
}
static void __iomem *omapfb_get_region_vaddr(const struct omapfb_info *ofbi)
{
if (ofbi->rotation_type == OMAP_DSS_ROT_VRFB)
return ofbi->region->vrfb.vaddr[0];
else
return ofbi->region->vaddr;
}
static struct omapfb_colormode omapfb_colormodes[] = {
{
.dssmode = OMAP_DSS_COLOR_UYVY,
.bits_per_pixel = 16,
.nonstd = OMAPFB_COLOR_YUV422,
}, {
.dssmode = OMAP_DSS_COLOR_YUV2,
.bits_per_pixel = 16,
.nonstd = OMAPFB_COLOR_YUY422,
}, {
.dssmode = OMAP_DSS_COLOR_ARGB16,
.bits_per_pixel = 16,
.red = { .length = 4, .offset = 8, .msb_right = 0 },
.green = { .length = 4, .offset = 4, .msb_right = 0 },
.blue = { .length = 4, .offset = 0, .msb_right = 0 },
.transp = { .length = 4, .offset = 12, .msb_right = 0 },
}, {
.dssmode = OMAP_DSS_COLOR_RGB16,
.bits_per_pixel = 16,
.red = { .length = 5, .offset = 11, .msb_right = 0 },
.green = { .length = 6, .offset = 5, .msb_right = 0 },
.blue = { .length = 5, .offset = 0, .msb_right = 0 },
.transp = { .length = 0, .offset = 0, .msb_right = 0 },
}, {
.dssmode = OMAP_DSS_COLOR_RGB24P,
.bits_per_pixel = 24,
.red = { .length = 8, .offset = 16, .msb_right = 0 },
.green = { .length = 8, .offset = 8, .msb_right = 0 },
.blue = { .length = 8, .offset = 0, .msb_right = 0 },
.transp = { .length = 0, .offset = 0, .msb_right = 0 },
}, {
.dssmode = OMAP_DSS_COLOR_RGB24U,
.bits_per_pixel = 32,
.red = { .length = 8, .offset = 16, .msb_right = 0 },
.green = { .length = 8, .offset = 8, .msb_right = 0 },
.blue = { .length = 8, .offset = 0, .msb_right = 0 },
.transp = { .length = 0, .offset = 0, .msb_right = 0 },
}, {
.dssmode = OMAP_DSS_COLOR_ARGB32,
.bits_per_pixel = 32,
.red = { .length = 8, .offset = 16, .msb_right = 0 },
.green = { .length = 8, .offset = 8, .msb_right = 0 },
.blue = { .length = 8, .offset = 0, .msb_right = 0 },
.transp = { .length = 8, .offset = 24, .msb_right = 0 },
}, {
.dssmode = OMAP_DSS_COLOR_RGBA32,
.bits_per_pixel = 32,
.red = { .length = 8, .offset = 24, .msb_right = 0 },
.green = { .length = 8, .offset = 16, .msb_right = 0 },
.blue = { .length = 8, .offset = 8, .msb_right = 0 },
.transp = { .length = 8, .offset = 0, .msb_right = 0 },
}, {
.dssmode = OMAP_DSS_COLOR_RGBX32,
.bits_per_pixel = 32,
.red = { .length = 8, .offset = 24, .msb_right = 0 },
.green = { .length = 8, .offset = 16, .msb_right = 0 },
.blue = { .length = 8, .offset = 8, .msb_right = 0 },
.transp = { .length = 0, .offset = 0, .msb_right = 0 },
},
};
static bool cmp_component(struct fb_bitfield *f1, struct fb_bitfield *f2)
{
return f1->length == f2->length &&
f1->offset == f2->offset &&
f1->msb_right == f2->msb_right;
}
static bool cmp_var_to_colormode(struct fb_var_screeninfo *var,
struct omapfb_colormode *color)
{
if (var->bits_per_pixel == 0 ||
var->red.length == 0 ||
var->blue.length == 0 ||
var->green.length == 0)
return false;
return var->bits_per_pixel == color->bits_per_pixel &&
cmp_component(&var->red, &color->red) &&
cmp_component(&var->green, &color->green) &&
cmp_component(&var->blue, &color->blue) &&
cmp_component(&var->transp, &color->transp);
}
static void assign_colormode_to_var(struct fb_var_screeninfo *var,
struct omapfb_colormode *color)
{
var->bits_per_pixel = color->bits_per_pixel;
var->nonstd = color->nonstd;
var->red = color->red;
var->green = color->green;
var->blue = color->blue;
var->transp = color->transp;
}
static int fb_mode_to_dss_mode(struct fb_var_screeninfo *var,
enum omap_color_mode *mode)
{
enum omap_color_mode dssmode;
int i;
/* first match with nonstd field */
if (var->nonstd) {
for (i = 0; i < ARRAY_SIZE(omapfb_colormodes); ++i) {
struct omapfb_colormode *m = &omapfb_colormodes[i];
if (var->nonstd == m->nonstd) {
assign_colormode_to_var(var, m);
*mode = m->dssmode;
return 0;
}
}
return -EINVAL;
}
/* then try exact match of bpp and colors */
for (i = 0; i < ARRAY_SIZE(omapfb_colormodes); ++i) {
struct omapfb_colormode *m = &omapfb_colormodes[i];
if (cmp_var_to_colormode(var, m)) {
assign_colormode_to_var(var, m);
*mode = m->dssmode;
return 0;
}
}
/* match with bpp if user has not filled color fields
* properly */
switch (var->bits_per_pixel) {
case 1:
dssmode = OMAP_DSS_COLOR_CLUT1;
break;
case 2:
dssmode = OMAP_DSS_COLOR_CLUT2;
break;
case 4:
dssmode = OMAP_DSS_COLOR_CLUT4;
break;
case 8:
dssmode = OMAP_DSS_COLOR_CLUT8;
break;
case 12:
dssmode = OMAP_DSS_COLOR_RGB12U;
break;
case 16:
dssmode = OMAP_DSS_COLOR_RGB16;
break;
case 24:
dssmode = OMAP_DSS_COLOR_RGB24P;
break;
case 32:
dssmode = OMAP_DSS_COLOR_RGB24U;
break;
default:
return -EINVAL;
}
for (i = 0; i < ARRAY_SIZE(omapfb_colormodes); ++i) {
struct omapfb_colormode *m = &omapfb_colormodes[i];
if (dssmode == m->dssmode) {
assign_colormode_to_var(var, m);
*mode = m->dssmode;
return 0;
}
}
return -EINVAL;
}
static int check_fb_res_bounds(struct fb_var_screeninfo *var)
{
int xres_min = OMAPFB_PLANE_XRES_MIN;
int xres_max = 2048;
int yres_min = OMAPFB_PLANE_YRES_MIN;
int yres_max = 2048;
/* XXX: some applications seem to set virtual res to 0. */
if (var->xres_virtual == 0)
var->xres_virtual = var->xres;
if (var->yres_virtual == 0)
var->yres_virtual = var->yres;
if (var->xres_virtual < xres_min || var->yres_virtual < yres_min)
return -EINVAL;
if (var->xres < xres_min)
var->xres = xres_min;
if (var->yres < yres_min)
var->yres = yres_min;
if (var->xres > xres_max)
var->xres = xres_max;
if (var->yres > yres_max)
var->yres = yres_max;
if (var->xres > var->xres_virtual)
var->xres = var->xres_virtual;
if (var->yres > var->yres_virtual)
var->yres = var->yres_virtual;
return 0;
}
static void shrink_height(unsigned long max_frame_size,
struct fb_var_screeninfo *var)
{
DBG("can't fit FB into memory, reducing y\n");
var->yres_virtual = max_frame_size /
(var->xres_virtual * var->bits_per_pixel >> 3);
if (var->yres_virtual < OMAPFB_PLANE_YRES_MIN)
var->yres_virtual = OMAPFB_PLANE_YRES_MIN;
if (var->yres > var->yres_virtual)
var->yres = var->yres_virtual;
}
static void shrink_width(unsigned long max_frame_size,
struct fb_var_screeninfo *var)
{
DBG("can't fit FB into memory, reducing x\n");
var->xres_virtual = max_frame_size / var->yres_virtual /
(var->bits_per_pixel >> 3);
if (var->xres_virtual < OMAPFB_PLANE_XRES_MIN)
var->xres_virtual = OMAPFB_PLANE_XRES_MIN;
if (var->xres > var->xres_virtual)
var->xres = var->xres_virtual;
}
static int check_vrfb_fb_size(unsigned long region_size,
const struct fb_var_screeninfo *var)
{
unsigned long min_phys_size = omap_vrfb_min_phys_size(var->xres_virtual,
var->yres_virtual, var->bits_per_pixel >> 3);
return min_phys_size > region_size ? -EINVAL : 0;
}
static int check_fb_size(const struct omapfb_info *ofbi,
struct fb_var_screeninfo *var)
{
unsigned long max_frame_size = ofbi->region->size;
int bytespp = var->bits_per_pixel >> 3;
unsigned long line_size = var->xres_virtual * bytespp;
if (ofbi->rotation_type == OMAP_DSS_ROT_VRFB) {
/* One needs to check for both VRFB and OMAPFB limitations. */
if (check_vrfb_fb_size(max_frame_size, var))
shrink_height(omap_vrfb_max_height(
max_frame_size, var->xres_virtual, bytespp) *
line_size, var);
if (check_vrfb_fb_size(max_frame_size, var)) {
DBG("cannot fit FB to memory\n");
return -EINVAL;
}
return 0;
}
DBG("max frame size %lu, line size %lu\n", max_frame_size, line_size);
if (line_size * var->yres_virtual > max_frame_size)
shrink_height(max_frame_size, var);
if (line_size * var->yres_virtual > max_frame_size) {
shrink_width(max_frame_size, var);
line_size = var->xres_virtual * bytespp;
}
if (line_size * var->yres_virtual > max_frame_size) {
DBG("cannot fit FB to memory\n");
return -EINVAL;
}
return 0;
}
/*
* Consider if VRFB assisted rotation is in use and if the virtual space for
* the zero degree view needs to be mapped. The need for mapping also acts as
* the trigger for setting up the hardware on the context in question. This
* ensures that one does not attempt to access the virtual view before the
* hardware is serving the address translations.
*/
static int setup_vrfb_rotation(struct fb_info *fbi)
{
struct omapfb_info *ofbi = FB2OFB(fbi);
struct omapfb2_mem_region *rg = ofbi->region;
struct vrfb *vrfb = &rg->vrfb;
struct fb_var_screeninfo *var = &fbi->var;
struct fb_fix_screeninfo *fix = &fbi->fix;
unsigned bytespp;
bool yuv_mode;
enum omap_color_mode mode;
int r;
bool reconf;
if (!rg->size || ofbi->rotation_type != OMAP_DSS_ROT_VRFB)
return 0;
DBG("setup_vrfb_rotation\n");
r = fb_mode_to_dss_mode(var, &mode);
if (r)
return r;
bytespp = var->bits_per_pixel >> 3;
yuv_mode = mode == OMAP_DSS_COLOR_YUV2 || mode == OMAP_DSS_COLOR_UYVY;
/* We need to reconfigure VRFB if the resolution changes, if yuv mode
* is enabled/disabled, or if bytes per pixel changes */
/* XXX we shouldn't allow this when framebuffer is mmapped */
reconf = false;
if (yuv_mode != vrfb->yuv_mode)
reconf = true;
else if (bytespp != vrfb->bytespp)
reconf = true;
else if (vrfb->xres != var->xres_virtual ||
vrfb->yres != var->yres_virtual)
reconf = true;
if (vrfb->vaddr[0] && reconf) {
fbi->screen_base = NULL;
fix->smem_start = 0;
fix->smem_len = 0;
iounmap(vrfb->vaddr[0]);
vrfb->vaddr[0] = NULL;
DBG("setup_vrfb_rotation: reset fb\n");
}
if (vrfb->vaddr[0])
return 0;
omap_vrfb_setup(&rg->vrfb, rg->paddr,
var->xres_virtual,
var->yres_virtual,
bytespp, yuv_mode);
/* Now one can ioremap the 0 angle view */
r = omap_vrfb_map_angle(vrfb, var->yres_virtual, 0);
if (r)
return r;
/* used by open/write in fbmem.c */
fbi->screen_base = ofbi->region->vrfb.vaddr[0];
fix->smem_start = ofbi->region->vrfb.paddr[0];
switch (var->nonstd) {
case OMAPFB_COLOR_YUV422:
case OMAPFB_COLOR_YUY422:
fix->line_length =
(OMAP_VRFB_LINE_LEN * var->bits_per_pixel) >> 2;
break;
default:
fix->line_length =
(OMAP_VRFB_LINE_LEN * var->bits_per_pixel) >> 3;
break;
}
fix->smem_len = var->yres_virtual * fix->line_length;
return 0;
}
int dss_mode_to_fb_mode(enum omap_color_mode dssmode,
struct fb_var_screeninfo *var)
{
int i;
for (i = 0; i < ARRAY_SIZE(omapfb_colormodes); ++i) {
struct omapfb_colormode *mode = &omapfb_colormodes[i];
if (dssmode == mode->dssmode) {
assign_colormode_to_var(var, mode);
return 0;
}
}
return -ENOENT;
}
void set_fb_fix(struct fb_info *fbi)
{
struct fb_fix_screeninfo *fix = &fbi->fix;
struct fb_var_screeninfo *var = &fbi->var;
struct omapfb_info *ofbi = FB2OFB(fbi);
struct omapfb2_mem_region *rg = ofbi->region;
DBG("set_fb_fix\n");
/* used by open/write in fbmem.c */
fbi->screen_base = (char __iomem *)omapfb_get_region_vaddr(ofbi);
/* used by mmap in fbmem.c */
if (ofbi->rotation_type == OMAP_DSS_ROT_VRFB) {
switch (var->nonstd) {
case OMAPFB_COLOR_YUV422:
case OMAPFB_COLOR_YUY422:
fix->line_length =
(OMAP_VRFB_LINE_LEN * var->bits_per_pixel) >> 2;
break;
default:
fix->line_length =
(OMAP_VRFB_LINE_LEN * var->bits_per_pixel) >> 3;
break;
}
fix->smem_len = var->yres_virtual * fix->line_length;
} else {
fix->line_length =
(var->xres_virtual * var->bits_per_pixel) >> 3;
fix->smem_len = rg->size;
}
fix->smem_start = omapfb_get_region_paddr(ofbi);
fix->type = FB_TYPE_PACKED_PIXELS;
if (var->nonstd)
fix->visual = FB_VISUAL_PSEUDOCOLOR;
else {
switch (var->bits_per_pixel) {
case 32:
case 24:
case 16:
case 12:
fix->visual = FB_VISUAL_TRUECOLOR;
/* 12bpp is stored in 16 bits */
break;
case 1:
case 2:
case 4:
case 8:
fix->visual = FB_VISUAL_PSEUDOCOLOR;
break;
}
}
fix->accel = FB_ACCEL_NONE;
fix->xpanstep = 1;
fix->ypanstep = 1;
}
/* check new var and possibly modify it to be ok */
int check_fb_var(struct fb_info *fbi, struct fb_var_screeninfo *var)
{
struct omapfb_info *ofbi = FB2OFB(fbi);
struct omap_dss_device *display = fb2display(fbi);
enum omap_color_mode mode = 0;
int i;
int r;
DBG("check_fb_var %d\n", ofbi->id);
WARN_ON(!atomic_read(&ofbi->region->lock_count));
r = fb_mode_to_dss_mode(var, &mode);
if (r) {
DBG("cannot convert var to omap dss mode\n");
return r;
}
for (i = 0; i < ofbi->num_overlays; ++i) {
if ((ofbi->overlays[i]->supported_modes & mode) == 0) {
DBG("invalid mode\n");
return -EINVAL;
}
}
if (var->rotate > 3)
return -EINVAL;
if (check_fb_res_bounds(var))
return -EINVAL;
/* When no memory is allocated ignore the size check */
if (ofbi->region->size != 0 && check_fb_size(ofbi, var))
return -EINVAL;
if (var->xres + var->xoffset > var->xres_virtual)
var->xoffset = var->xres_virtual - var->xres;
if (var->yres + var->yoffset > var->yres_virtual)
var->yoffset = var->yres_virtual - var->yres;
DBG("xres = %d, yres = %d, vxres = %d, vyres = %d\n",
var->xres, var->yres,
var->xres_virtual, var->yres_virtual);
if (display && display->driver->get_dimensions) {
u32 w, h;
display->driver->get_dimensions(display, &w, &h);
var->width = DIV_ROUND_CLOSEST(w, 1000);
var->height = DIV_ROUND_CLOSEST(h, 1000);
} else {
var->height = -1;
var->width = -1;
}
var->grayscale = 0;
if (display && display->driver->get_timings) {
struct omap_video_timings timings;
display->driver->get_timings(display, &timings);
/* pixclock in ps, the rest in pixclock */
var->pixclock = timings.pixelclock != 0 ?
KHZ2PICOS(timings.pixelclock / 1000) :
0;
var->left_margin = timings.hbp;
var->right_margin = timings.hfp;
var->upper_margin = timings.vbp;
var->lower_margin = timings.vfp;
var->hsync_len = timings.hsw;
var->vsync_len = timings.vsw;
var->sync |= timings.hsync_level == OMAPDSS_SIG_ACTIVE_HIGH ?
FB_SYNC_HOR_HIGH_ACT : 0;
var->sync |= timings.vsync_level == OMAPDSS_SIG_ACTIVE_HIGH ?
FB_SYNC_VERT_HIGH_ACT : 0;
var->vmode = timings.interlace ?
FB_VMODE_INTERLACED : FB_VMODE_NONINTERLACED;
} else {
var->pixclock = 0;
var->left_margin = 0;
var->right_margin = 0;
var->upper_margin = 0;
var->lower_margin = 0;
var->hsync_len = 0;
var->vsync_len = 0;
var->sync = 0;
var->vmode = FB_VMODE_NONINTERLACED;
}
return 0;
}
/*
* ---------------------------------------------------------------------------
* fbdev framework callbacks
* ---------------------------------------------------------------------------
*/
static int omapfb_open(struct fb_info *fbi, int user)
{
return 0;
}
static int omapfb_release(struct fb_info *fbi, int user)
{
return 0;
}
static unsigned calc_rotation_offset_dma(const struct fb_var_screeninfo *var,
const struct fb_fix_screeninfo *fix, int rotation)
{
unsigned offset;
offset = var->yoffset * fix->line_length +
var->xoffset * (var->bits_per_pixel >> 3);
return offset;
}
static unsigned calc_rotation_offset_vrfb(const struct fb_var_screeninfo *var,
const struct fb_fix_screeninfo *fix, int rotation)
{
unsigned offset;
if (rotation == FB_ROTATE_UD)
offset = (var->yres_virtual - var->yres) *
fix->line_length;
else if (rotation == FB_ROTATE_CW)
offset = (var->yres_virtual - var->yres) *
(var->bits_per_pixel >> 3);
else
offset = 0;
if (rotation == FB_ROTATE_UR)
offset += var->yoffset * fix->line_length +
var->xoffset * (var->bits_per_pixel >> 3);
else if (rotation == FB_ROTATE_UD)
offset -= var->yoffset * fix->line_length +
var->xoffset * (var->bits_per_pixel >> 3);
else if (rotation == FB_ROTATE_CW)
offset -= var->xoffset * fix->line_length +
var->yoffset * (var->bits_per_pixel >> 3);
else if (rotation == FB_ROTATE_CCW)
offset += var->xoffset * fix->line_length +
var->yoffset * (var->bits_per_pixel >> 3);
return offset;
}
static void omapfb_calc_addr(const struct omapfb_info *ofbi,
const struct fb_var_screeninfo *var,
const struct fb_fix_screeninfo *fix,
int rotation, u32 *paddr)
{
u32 data_start_p;
int offset;
if (ofbi->rotation_type == OMAP_DSS_ROT_VRFB)
data_start_p = omapfb_get_region_rot_paddr(ofbi, rotation);
else
data_start_p = omapfb_get_region_paddr(ofbi);
if (ofbi->rotation_type == OMAP_DSS_ROT_VRFB)
offset = calc_rotation_offset_vrfb(var, fix, rotation);
else
offset = calc_rotation_offset_dma(var, fix, rotation);
data_start_p += offset;
if (offset)
DBG("offset %d, %d = %d\n",
var->xoffset, var->yoffset, offset);
DBG("paddr %x\n", data_start_p);
*paddr = data_start_p;
}
/* setup overlay according to the fb */
int omapfb_setup_overlay(struct fb_info *fbi, struct omap_overlay *ovl,
u16 posx, u16 posy, u16 outw, u16 outh)
{
int r = 0;
struct omapfb_info *ofbi = FB2OFB(fbi);
struct fb_var_screeninfo *var = &fbi->var;
struct fb_fix_screeninfo *fix = &fbi->fix;
enum omap_color_mode mode = 0;
u32 data_start_p = 0;
struct omap_overlay_info info;
int xres, yres;
int screen_width;
int mirror;
int rotation = var->rotate;
int i;
WARN_ON(!atomic_read(&ofbi->region->lock_count));
for (i = 0; i < ofbi->num_overlays; i++) {
if (ovl != ofbi->overlays[i])
continue;
rotation = (rotation + ofbi->rotation[i]) % 4;
break;
}
DBG("setup_overlay %d, posx %d, posy %d, outw %d, outh %d\n", ofbi->id,
posx, posy, outw, outh);
if (rotation == FB_ROTATE_CW || rotation == FB_ROTATE_CCW) {
xres = var->yres;
yres = var->xres;
} else {
xres = var->xres;
yres = var->yres;
}
if (ofbi->region->size)
omapfb_calc_addr(ofbi, var, fix, rotation, &data_start_p);
r = fb_mode_to_dss_mode(var, &mode);
if (r) {
DBG("fb_mode_to_dss_mode failed");
goto err;
}
switch (var->nonstd) {
case OMAPFB_COLOR_YUV422:
case OMAPFB_COLOR_YUY422:
if (ofbi->rotation_type == OMAP_DSS_ROT_VRFB) {
screen_width = fix->line_length
/ (var->bits_per_pixel >> 2);
break;
}
/* fall through */
default:
screen_width = fix->line_length / (var->bits_per_pixel >> 3);
break;
}
ovl->get_overlay_info(ovl, &info);
if (ofbi->rotation_type == OMAP_DSS_ROT_VRFB)
mirror = 0;
else
mirror = ofbi->mirror;
info.paddr = data_start_p;
info.screen_width = screen_width;
info.width = xres;
info.height = yres;
info.color_mode = mode;
info.rotation_type = ofbi->rotation_type;
info.rotation = rotation;
info.mirror = mirror;
info.pos_x = posx;
info.pos_y = posy;
info.out_width = outw;
info.out_height = outh;
r = ovl->set_overlay_info(ovl, &info);
if (r) {
DBG("ovl->setup_overlay_info failed\n");
goto err;
}
return 0;
err:
DBG("setup_overlay failed\n");
return r;
}
/* apply var to the overlay */
int omapfb_apply_changes(struct fb_info *fbi, int init)
{
int r = 0;
struct omapfb_info *ofbi = FB2OFB(fbi);
struct fb_var_screeninfo *var = &fbi->var;
struct omap_overlay *ovl;
u16 posx, posy;
u16 outw, outh;
int i;
#ifdef DEBUG
if (omapfb_test_pattern)
fill_fb(fbi);
#endif
WARN_ON(!atomic_read(&ofbi->region->lock_count));
for (i = 0; i < ofbi->num_overlays; i++) {
ovl = ofbi->overlays[i];
DBG("apply_changes, fb %d, ovl %d\n", ofbi->id, ovl->id);
if (ofbi->region->size == 0) {
/* the fb is not available. disable the overlay */
omapfb_overlay_enable(ovl, 0);
if (!init && ovl->manager)
ovl->manager->apply(ovl->manager);
continue;
}
if (init || (ovl->caps & OMAP_DSS_OVL_CAP_SCALE) == 0) {
int rotation = (var->rotate + ofbi->rotation[i]) % 4;
if (rotation == FB_ROTATE_CW ||
rotation == FB_ROTATE_CCW) {
outw = var->yres;
outh = var->xres;
} else {
outw = var->xres;
outh = var->yres;
}
} else {
struct omap_overlay_info info;
ovl->get_overlay_info(ovl, &info);
outw = info.out_width;
outh = info.out_height;
}
if (init) {
posx = 0;
posy = 0;
} else {
struct omap_overlay_info info;
ovl->get_overlay_info(ovl, &info);
posx = info.pos_x;
posy = info.pos_y;
}
r = omapfb_setup_overlay(fbi, ovl, posx, posy, outw, outh);
if (r)
goto err;
if (!init && ovl->manager)
ovl->manager->apply(ovl->manager);
}
return 0;
err:
DBG("apply_changes failed\n");
return r;
}
/* checks var and eventually tweaks it to something supported,
* DO NOT MODIFY PAR */
static int omapfb_check_var(struct fb_var_screeninfo *var, struct fb_info *fbi)
{
struct omapfb_info *ofbi = FB2OFB(fbi);
int r;
DBG("check_var(%d)\n", FB2OFB(fbi)->id);
omapfb_get_mem_region(ofbi->region);
r = check_fb_var(fbi, var);
omapfb_put_mem_region(ofbi->region);
return r;
}
/* set the video mode according to info->var */
static int omapfb_set_par(struct fb_info *fbi)
{
struct omapfb_info *ofbi = FB2OFB(fbi);
int r;
DBG("set_par(%d)\n", FB2OFB(fbi)->id);
omapfb_get_mem_region(ofbi->region);
set_fb_fix(fbi);
r = setup_vrfb_rotation(fbi);
if (r)
goto out;
r = omapfb_apply_changes(fbi, 0);
out:
omapfb_put_mem_region(ofbi->region);
return r;
}
static int omapfb_pan_display(struct fb_var_screeninfo *var,
struct fb_info *fbi)
{
struct omapfb_info *ofbi = FB2OFB(fbi);
struct fb_var_screeninfo new_var;
int r;
DBG("pan_display(%d)\n", FB2OFB(fbi)->id);
if (var->xoffset == fbi->var.xoffset &&
var->yoffset == fbi->var.yoffset)
return 0;
new_var = fbi->var;
new_var.xoffset = var->xoffset;
new_var.yoffset = var->yoffset;
fbi->var = new_var;
omapfb_get_mem_region(ofbi->region);
r = omapfb_apply_changes(fbi, 0);
omapfb_put_mem_region(ofbi->region);
return r;
}
static void mmap_user_open(struct vm_area_struct *vma)
{
struct omapfb2_mem_region *rg = vma->vm_private_data;
omapfb_get_mem_region(rg);
atomic_inc(&rg->map_count);
omapfb_put_mem_region(rg);
}
static void mmap_user_close(struct vm_area_struct *vma)
{
struct omapfb2_mem_region *rg = vma->vm_private_data;
omapfb_get_mem_region(rg);
atomic_dec(&rg->map_count);
omapfb_put_mem_region(rg);
}
static const struct vm_operations_struct mmap_user_ops = {
.open = mmap_user_open,
.close = mmap_user_close,
};
static int omapfb_mmap(struct fb_info *fbi, struct vm_area_struct *vma)
{
struct omapfb_info *ofbi = FB2OFB(fbi);
struct fb_fix_screeninfo *fix = &fbi->fix;
struct omapfb2_mem_region *rg;
unsigned long start;
u32 len;
int r;
rg = omapfb_get_mem_region(ofbi->region);
start = omapfb_get_region_paddr(ofbi);
len = fix->smem_len;
DBG("user mmap region start %lx, len %d, off %lx\n", start, len,
vma->vm_pgoff << PAGE_SHIFT);
vma->vm_page_prot = pgprot_writecombine(vma->vm_page_prot);
vma->vm_ops = &mmap_user_ops;
vma->vm_private_data = rg;
r = vm_iomap_memory(vma, start, len);
if (r)
goto error;
/* vm_ops.open won't be called for mmap itself. */
atomic_inc(&rg->map_count);
omapfb_put_mem_region(rg);
return 0;
error:
omapfb_put_mem_region(ofbi->region);
return r;
}
/* Store a single color palette entry into a pseudo palette or the hardware
* palette if one is available. For now we support only 16bpp and thus store
* the entry only to the pseudo palette.
*/
static int _setcolreg(struct fb_info *fbi, u_int regno, u_int red, u_int green,
u_int blue, u_int transp, int update_hw_pal)
{
/*struct omapfb_info *ofbi = FB2OFB(fbi);*/
/*struct omapfb2_device *fbdev = ofbi->fbdev;*/
struct fb_var_screeninfo *var = &fbi->var;
int r = 0;
enum omapfb_color_format mode = OMAPFB_COLOR_RGB24U; /* XXX */
/*switch (plane->color_mode) {*/
switch (mode) {
case OMAPFB_COLOR_YUV422:
case OMAPFB_COLOR_YUV420:
case OMAPFB_COLOR_YUY422:
r = -EINVAL;
break;
case OMAPFB_COLOR_CLUT_8BPP:
case OMAPFB_COLOR_CLUT_4BPP:
case OMAPFB_COLOR_CLUT_2BPP:
case OMAPFB_COLOR_CLUT_1BPP:
/*
if (fbdev->ctrl->setcolreg)
r = fbdev->ctrl->setcolreg(regno, red, green, blue,
transp, update_hw_pal);
*/
/* Fallthrough */
r = -EINVAL;
break;
case OMAPFB_COLOR_RGB565:
case OMAPFB_COLOR_RGB444:
case OMAPFB_COLOR_RGB24P:
case OMAPFB_COLOR_RGB24U:
if (r != 0)
break;
if (regno < 16) {
u32 pal;
pal = ((red >> (16 - var->red.length)) <<
var->red.offset) |
((green >> (16 - var->green.length)) <<
var->green.offset) |
(blue >> (16 - var->blue.length));
((u32 *)(fbi->pseudo_palette))[regno] = pal;
}
break;
default:
BUG();
}
return r;
}
static int omapfb_setcolreg(u_int regno, u_int red, u_int green, u_int blue,
u_int transp, struct fb_info *info)
{
DBG("setcolreg\n");
return _setcolreg(info, regno, red, green, blue, transp, 1);
}
static int omapfb_setcmap(struct fb_cmap *cmap, struct fb_info *info)
{
int count, index, r;
u16 *red, *green, *blue, *transp;
u16 trans = 0xffff;
DBG("setcmap\n");
red = cmap->red;
green = cmap->green;
blue = cmap->blue;
transp = cmap->transp;
index = cmap->start;
for (count = 0; count < cmap->len; count++) {
if (transp)
trans = *transp++;
r = _setcolreg(info, index++, *red++, *green++, *blue++, trans,
count == cmap->len - 1);
if (r != 0)
return r;
}
return 0;
}
static int omapfb_blank(int blank, struct fb_info *fbi)
{
struct omapfb_info *ofbi = FB2OFB(fbi);
struct omapfb2_device *fbdev = ofbi->fbdev;
struct omap_dss_device *display = fb2display(fbi);
struct omapfb_display_data *d;
int r = 0;
if (!display)
return -EINVAL;
omapfb_lock(fbdev);
d = get_display_data(fbdev, display);
switch (blank) {
case FB_BLANK_UNBLANK:
if (display->state == OMAP_DSS_DISPLAY_ACTIVE)
goto exit;
r = display->driver->enable(display);
if ((display->caps & OMAP_DSS_DISPLAY_CAP_MANUAL_UPDATE) &&
d->update_mode == OMAPFB_AUTO_UPDATE &&
!d->auto_update_work_enabled)
omapfb_start_auto_update(fbdev, display);
break;
case FB_BLANK_NORMAL:
/* FB_BLANK_NORMAL could be implemented.
* Needs DSS additions. */
case FB_BLANK_VSYNC_SUSPEND:
case FB_BLANK_HSYNC_SUSPEND:
case FB_BLANK_POWERDOWN:
if (display->state != OMAP_DSS_DISPLAY_ACTIVE)
goto exit;
if (d->auto_update_work_enabled)
omapfb_stop_auto_update(fbdev, display);
display->driver->disable(display);
break;
default:
r = -EINVAL;
}
exit:
omapfb_unlock(fbdev);
return r;
}
#if 0
/* XXX fb_read and fb_write are needed for VRFB */
ssize_t omapfb_write(struct fb_info *info, const char __user *buf,
size_t count, loff_t *ppos)
{
DBG("omapfb_write %d, %lu\n", count, (unsigned long)*ppos);
/* XXX needed for VRFB */
return count;
}
#endif
static struct fb_ops omapfb_ops = {
.owner = THIS_MODULE,
.fb_open = omapfb_open,
.fb_release = omapfb_release,
.fb_fillrect = cfb_fillrect,
.fb_copyarea = cfb_copyarea,
.fb_imageblit = cfb_imageblit,
.fb_blank = omapfb_blank,
.fb_ioctl = omapfb_ioctl,
.fb_check_var = omapfb_check_var,
.fb_set_par = omapfb_set_par,
.fb_pan_display = omapfb_pan_display,
.fb_mmap = omapfb_mmap,
.fb_setcolreg = omapfb_setcolreg,
.fb_setcmap = omapfb_setcmap,
/*.fb_write = omapfb_write,*/
};
static void omapfb_free_fbmem(struct fb_info *fbi)
{
struct omapfb_info *ofbi = FB2OFB(fbi);
struct omapfb2_device *fbdev = ofbi->fbdev;
struct omapfb2_mem_region *rg;
rg = ofbi->region;
if (rg->token == NULL)
return;
WARN_ON(atomic_read(&rg->map_count));
if (ofbi->rotation_type == OMAP_DSS_ROT_VRFB) {
/* unmap the 0 angle rotation */
if (rg->vrfb.vaddr[0]) {
iounmap(rg->vrfb.vaddr[0]);
rg->vrfb.vaddr[0] = NULL;
}
omap_vrfb_release_ctx(&rg->vrfb);
}
dma_free_attrs(fbdev->dev, rg->size, rg->token, rg->dma_handle,
dma-mapping: use unsigned long for dma_attrs The dma-mapping core and the implementations do not change the DMA attributes passed by pointer. Thus the pointer can point to const data. However the attributes do not have to be a bitfield. Instead unsigned long will do fine: 1. This is just simpler. Both in terms of reading the code and setting attributes. Instead of initializing local attributes on the stack and passing pointer to it to dma_set_attr(), just set the bits. 2. It brings safeness and checking for const correctness because the attributes are passed by value. Semantic patches for this change (at least most of them): virtual patch virtual context @r@ identifier f, attrs; @@ f(..., - struct dma_attrs *attrs + unsigned long attrs , ...) { ... } @@ identifier r.f; @@ f(..., - NULL + 0 ) and // Options: --all-includes virtual patch virtual context @r@ identifier f, attrs; type t; @@ t f(..., struct dma_attrs *attrs); @@ identifier r.f; @@ f(..., - NULL + 0 ) Link: http://lkml.kernel.org/r/1468399300-5399-2-git-send-email-k.kozlowski@samsung.com Signed-off-by: Krzysztof Kozlowski <k.kozlowski@samsung.com> Acked-by: Vineet Gupta <vgupta@synopsys.com> Acked-by: Robin Murphy <robin.murphy@arm.com> Acked-by: Hans-Christian Noren Egtvedt <egtvedt@samfundet.no> Acked-by: Mark Salter <msalter@redhat.com> [c6x] Acked-by: Jesper Nilsson <jesper.nilsson@axis.com> [cris] Acked-by: Daniel Vetter <daniel.vetter@ffwll.ch> [drm] Reviewed-by: Bart Van Assche <bart.vanassche@sandisk.com> Acked-by: Joerg Roedel <jroedel@suse.de> [iommu] Acked-by: Fabien Dessenne <fabien.dessenne@st.com> [bdisp] Reviewed-by: Marek Szyprowski <m.szyprowski@samsung.com> [vb2-core] Acked-by: David Vrabel <david.vrabel@citrix.com> [xen] Acked-by: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com> [xen swiotlb] Acked-by: Joerg Roedel <jroedel@suse.de> [iommu] Acked-by: Richard Kuo <rkuo@codeaurora.org> [hexagon] Acked-by: Geert Uytterhoeven <geert@linux-m68k.org> [m68k] Acked-by: Gerald Schaefer <gerald.schaefer@de.ibm.com> [s390] Acked-by: Bjorn Andersson <bjorn.andersson@linaro.org> Acked-by: Hans-Christian Noren Egtvedt <egtvedt@samfundet.no> [avr32] Acked-by: Vineet Gupta <vgupta@synopsys.com> [arc] Acked-by: Robin Murphy <robin.murphy@arm.com> [arm64 and dma-iommu] Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-08-03 14:46:00 -06:00
rg->attrs);
rg->token = NULL;
rg->vaddr = NULL;
rg->paddr = 0;
rg->alloc = 0;
rg->size = 0;
}
static void clear_fb_info(struct fb_info *fbi)
{
memset(&fbi->var, 0, sizeof(fbi->var));
memset(&fbi->fix, 0, sizeof(fbi->fix));
strlcpy(fbi->fix.id, MODULE_NAME, sizeof(fbi->fix.id));
}
static int omapfb_free_all_fbmem(struct omapfb2_device *fbdev)
{
int i;
DBG("free all fbmem\n");
for (i = 0; i < fbdev->num_fbs; i++) {
struct fb_info *fbi = fbdev->fbs[i];
omapfb_free_fbmem(fbi);
clear_fb_info(fbi);
}
return 0;
}
static int omapfb_alloc_fbmem(struct fb_info *fbi, unsigned long size,
unsigned long paddr)
{
struct omapfb_info *ofbi = FB2OFB(fbi);
struct omapfb2_device *fbdev = ofbi->fbdev;
struct omapfb2_mem_region *rg;
void *token;
dma-mapping: use unsigned long for dma_attrs The dma-mapping core and the implementations do not change the DMA attributes passed by pointer. Thus the pointer can point to const data. However the attributes do not have to be a bitfield. Instead unsigned long will do fine: 1. This is just simpler. Both in terms of reading the code and setting attributes. Instead of initializing local attributes on the stack and passing pointer to it to dma_set_attr(), just set the bits. 2. It brings safeness and checking for const correctness because the attributes are passed by value. Semantic patches for this change (at least most of them): virtual patch virtual context @r@ identifier f, attrs; @@ f(..., - struct dma_attrs *attrs + unsigned long attrs , ...) { ... } @@ identifier r.f; @@ f(..., - NULL + 0 ) and // Options: --all-includes virtual patch virtual context @r@ identifier f, attrs; type t; @@ t f(..., struct dma_attrs *attrs); @@ identifier r.f; @@ f(..., - NULL + 0 ) Link: http://lkml.kernel.org/r/1468399300-5399-2-git-send-email-k.kozlowski@samsung.com Signed-off-by: Krzysztof Kozlowski <k.kozlowski@samsung.com> Acked-by: Vineet Gupta <vgupta@synopsys.com> Acked-by: Robin Murphy <robin.murphy@arm.com> Acked-by: Hans-Christian Noren Egtvedt <egtvedt@samfundet.no> Acked-by: Mark Salter <msalter@redhat.com> [c6x] Acked-by: Jesper Nilsson <jesper.nilsson@axis.com> [cris] Acked-by: Daniel Vetter <daniel.vetter@ffwll.ch> [drm] Reviewed-by: Bart Van Assche <bart.vanassche@sandisk.com> Acked-by: Joerg Roedel <jroedel@suse.de> [iommu] Acked-by: Fabien Dessenne <fabien.dessenne@st.com> [bdisp] Reviewed-by: Marek Szyprowski <m.szyprowski@samsung.com> [vb2-core] Acked-by: David Vrabel <david.vrabel@citrix.com> [xen] Acked-by: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com> [xen swiotlb] Acked-by: Joerg Roedel <jroedel@suse.de> [iommu] Acked-by: Richard Kuo <rkuo@codeaurora.org> [hexagon] Acked-by: Geert Uytterhoeven <geert@linux-m68k.org> [m68k] Acked-by: Gerald Schaefer <gerald.schaefer@de.ibm.com> [s390] Acked-by: Bjorn Andersson <bjorn.andersson@linaro.org> Acked-by: Hans-Christian Noren Egtvedt <egtvedt@samfundet.no> [avr32] Acked-by: Vineet Gupta <vgupta@synopsys.com> [arc] Acked-by: Robin Murphy <robin.murphy@arm.com> [arm64 and dma-iommu] Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-08-03 14:46:00 -06:00
unsigned long attrs;
dma_addr_t dma_handle;
int r;
rg = ofbi->region;
rg->paddr = 0;
rg->vaddr = NULL;
memset(&rg->vrfb, 0, sizeof rg->vrfb);
rg->size = 0;
rg->type = 0;
rg->alloc = false;
rg->map = false;
size = PAGE_ALIGN(size);
dma-mapping: use unsigned long for dma_attrs The dma-mapping core and the implementations do not change the DMA attributes passed by pointer. Thus the pointer can point to const data. However the attributes do not have to be a bitfield. Instead unsigned long will do fine: 1. This is just simpler. Both in terms of reading the code and setting attributes. Instead of initializing local attributes on the stack and passing pointer to it to dma_set_attr(), just set the bits. 2. It brings safeness and checking for const correctness because the attributes are passed by value. Semantic patches for this change (at least most of them): virtual patch virtual context @r@ identifier f, attrs; @@ f(..., - struct dma_attrs *attrs + unsigned long attrs , ...) { ... } @@ identifier r.f; @@ f(..., - NULL + 0 ) and // Options: --all-includes virtual patch virtual context @r@ identifier f, attrs; type t; @@ t f(..., struct dma_attrs *attrs); @@ identifier r.f; @@ f(..., - NULL + 0 ) Link: http://lkml.kernel.org/r/1468399300-5399-2-git-send-email-k.kozlowski@samsung.com Signed-off-by: Krzysztof Kozlowski <k.kozlowski@samsung.com> Acked-by: Vineet Gupta <vgupta@synopsys.com> Acked-by: Robin Murphy <robin.murphy@arm.com> Acked-by: Hans-Christian Noren Egtvedt <egtvedt@samfundet.no> Acked-by: Mark Salter <msalter@redhat.com> [c6x] Acked-by: Jesper Nilsson <jesper.nilsson@axis.com> [cris] Acked-by: Daniel Vetter <daniel.vetter@ffwll.ch> [drm] Reviewed-by: Bart Van Assche <bart.vanassche@sandisk.com> Acked-by: Joerg Roedel <jroedel@suse.de> [iommu] Acked-by: Fabien Dessenne <fabien.dessenne@st.com> [bdisp] Reviewed-by: Marek Szyprowski <m.szyprowski@samsung.com> [vb2-core] Acked-by: David Vrabel <david.vrabel@citrix.com> [xen] Acked-by: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com> [xen swiotlb] Acked-by: Joerg Roedel <jroedel@suse.de> [iommu] Acked-by: Richard Kuo <rkuo@codeaurora.org> [hexagon] Acked-by: Geert Uytterhoeven <geert@linux-m68k.org> [m68k] Acked-by: Gerald Schaefer <gerald.schaefer@de.ibm.com> [s390] Acked-by: Bjorn Andersson <bjorn.andersson@linaro.org> Acked-by: Hans-Christian Noren Egtvedt <egtvedt@samfundet.no> [avr32] Acked-by: Vineet Gupta <vgupta@synopsys.com> [arc] Acked-by: Robin Murphy <robin.murphy@arm.com> [arm64 and dma-iommu] Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-08-03 14:46:00 -06:00
attrs = DMA_ATTR_WRITE_COMBINE;
if (ofbi->rotation_type == OMAP_DSS_ROT_VRFB)
dma-mapping: use unsigned long for dma_attrs The dma-mapping core and the implementations do not change the DMA attributes passed by pointer. Thus the pointer can point to const data. However the attributes do not have to be a bitfield. Instead unsigned long will do fine: 1. This is just simpler. Both in terms of reading the code and setting attributes. Instead of initializing local attributes on the stack and passing pointer to it to dma_set_attr(), just set the bits. 2. It brings safeness and checking for const correctness because the attributes are passed by value. Semantic patches for this change (at least most of them): virtual patch virtual context @r@ identifier f, attrs; @@ f(..., - struct dma_attrs *attrs + unsigned long attrs , ...) { ... } @@ identifier r.f; @@ f(..., - NULL + 0 ) and // Options: --all-includes virtual patch virtual context @r@ identifier f, attrs; type t; @@ t f(..., struct dma_attrs *attrs); @@ identifier r.f; @@ f(..., - NULL + 0 ) Link: http://lkml.kernel.org/r/1468399300-5399-2-git-send-email-k.kozlowski@samsung.com Signed-off-by: Krzysztof Kozlowski <k.kozlowski@samsung.com> Acked-by: Vineet Gupta <vgupta@synopsys.com> Acked-by: Robin Murphy <robin.murphy@arm.com> Acked-by: Hans-Christian Noren Egtvedt <egtvedt@samfundet.no> Acked-by: Mark Salter <msalter@redhat.com> [c6x] Acked-by: Jesper Nilsson <jesper.nilsson@axis.com> [cris] Acked-by: Daniel Vetter <daniel.vetter@ffwll.ch> [drm] Reviewed-by: Bart Van Assche <bart.vanassche@sandisk.com> Acked-by: Joerg Roedel <jroedel@suse.de> [iommu] Acked-by: Fabien Dessenne <fabien.dessenne@st.com> [bdisp] Reviewed-by: Marek Szyprowski <m.szyprowski@samsung.com> [vb2-core] Acked-by: David Vrabel <david.vrabel@citrix.com> [xen] Acked-by: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com> [xen swiotlb] Acked-by: Joerg Roedel <jroedel@suse.de> [iommu] Acked-by: Richard Kuo <rkuo@codeaurora.org> [hexagon] Acked-by: Geert Uytterhoeven <geert@linux-m68k.org> [m68k] Acked-by: Gerald Schaefer <gerald.schaefer@de.ibm.com> [s390] Acked-by: Bjorn Andersson <bjorn.andersson@linaro.org> Acked-by: Hans-Christian Noren Egtvedt <egtvedt@samfundet.no> [avr32] Acked-by: Vineet Gupta <vgupta@synopsys.com> [arc] Acked-by: Robin Murphy <robin.murphy@arm.com> [arm64 and dma-iommu] Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-08-03 14:46:00 -06:00
attrs |= DMA_ATTR_NO_KERNEL_MAPPING;
DBG("allocating %lu bytes for fb %d\n", size, ofbi->id);
token = dma_alloc_attrs(fbdev->dev, size, &dma_handle,
dma-mapping: use unsigned long for dma_attrs The dma-mapping core and the implementations do not change the DMA attributes passed by pointer. Thus the pointer can point to const data. However the attributes do not have to be a bitfield. Instead unsigned long will do fine: 1. This is just simpler. Both in terms of reading the code and setting attributes. Instead of initializing local attributes on the stack and passing pointer to it to dma_set_attr(), just set the bits. 2. It brings safeness and checking for const correctness because the attributes are passed by value. Semantic patches for this change (at least most of them): virtual patch virtual context @r@ identifier f, attrs; @@ f(..., - struct dma_attrs *attrs + unsigned long attrs , ...) { ... } @@ identifier r.f; @@ f(..., - NULL + 0 ) and // Options: --all-includes virtual patch virtual context @r@ identifier f, attrs; type t; @@ t f(..., struct dma_attrs *attrs); @@ identifier r.f; @@ f(..., - NULL + 0 ) Link: http://lkml.kernel.org/r/1468399300-5399-2-git-send-email-k.kozlowski@samsung.com Signed-off-by: Krzysztof Kozlowski <k.kozlowski@samsung.com> Acked-by: Vineet Gupta <vgupta@synopsys.com> Acked-by: Robin Murphy <robin.murphy@arm.com> Acked-by: Hans-Christian Noren Egtvedt <egtvedt@samfundet.no> Acked-by: Mark Salter <msalter@redhat.com> [c6x] Acked-by: Jesper Nilsson <jesper.nilsson@axis.com> [cris] Acked-by: Daniel Vetter <daniel.vetter@ffwll.ch> [drm] Reviewed-by: Bart Van Assche <bart.vanassche@sandisk.com> Acked-by: Joerg Roedel <jroedel@suse.de> [iommu] Acked-by: Fabien Dessenne <fabien.dessenne@st.com> [bdisp] Reviewed-by: Marek Szyprowski <m.szyprowski@samsung.com> [vb2-core] Acked-by: David Vrabel <david.vrabel@citrix.com> [xen] Acked-by: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com> [xen swiotlb] Acked-by: Joerg Roedel <jroedel@suse.de> [iommu] Acked-by: Richard Kuo <rkuo@codeaurora.org> [hexagon] Acked-by: Geert Uytterhoeven <geert@linux-m68k.org> [m68k] Acked-by: Gerald Schaefer <gerald.schaefer@de.ibm.com> [s390] Acked-by: Bjorn Andersson <bjorn.andersson@linaro.org> Acked-by: Hans-Christian Noren Egtvedt <egtvedt@samfundet.no> [avr32] Acked-by: Vineet Gupta <vgupta@synopsys.com> [arc] Acked-by: Robin Murphy <robin.murphy@arm.com> [arm64 and dma-iommu] Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-08-03 14:46:00 -06:00
GFP_KERNEL, attrs);
if (token == NULL) {
dev_err(fbdev->dev, "failed to allocate framebuffer\n");
return -ENOMEM;
}
DBG("allocated VRAM paddr %lx, vaddr %p\n",
(unsigned long)dma_handle, token);
if (ofbi->rotation_type == OMAP_DSS_ROT_VRFB) {
r = omap_vrfb_request_ctx(&rg->vrfb);
if (r) {
dma_free_attrs(fbdev->dev, size, token, dma_handle,
dma-mapping: use unsigned long for dma_attrs The dma-mapping core and the implementations do not change the DMA attributes passed by pointer. Thus the pointer can point to const data. However the attributes do not have to be a bitfield. Instead unsigned long will do fine: 1. This is just simpler. Both in terms of reading the code and setting attributes. Instead of initializing local attributes on the stack and passing pointer to it to dma_set_attr(), just set the bits. 2. It brings safeness and checking for const correctness because the attributes are passed by value. Semantic patches for this change (at least most of them): virtual patch virtual context @r@ identifier f, attrs; @@ f(..., - struct dma_attrs *attrs + unsigned long attrs , ...) { ... } @@ identifier r.f; @@ f(..., - NULL + 0 ) and // Options: --all-includes virtual patch virtual context @r@ identifier f, attrs; type t; @@ t f(..., struct dma_attrs *attrs); @@ identifier r.f; @@ f(..., - NULL + 0 ) Link: http://lkml.kernel.org/r/1468399300-5399-2-git-send-email-k.kozlowski@samsung.com Signed-off-by: Krzysztof Kozlowski <k.kozlowski@samsung.com> Acked-by: Vineet Gupta <vgupta@synopsys.com> Acked-by: Robin Murphy <robin.murphy@arm.com> Acked-by: Hans-Christian Noren Egtvedt <egtvedt@samfundet.no> Acked-by: Mark Salter <msalter@redhat.com> [c6x] Acked-by: Jesper Nilsson <jesper.nilsson@axis.com> [cris] Acked-by: Daniel Vetter <daniel.vetter@ffwll.ch> [drm] Reviewed-by: Bart Van Assche <bart.vanassche@sandisk.com> Acked-by: Joerg Roedel <jroedel@suse.de> [iommu] Acked-by: Fabien Dessenne <fabien.dessenne@st.com> [bdisp] Reviewed-by: Marek Szyprowski <m.szyprowski@samsung.com> [vb2-core] Acked-by: David Vrabel <david.vrabel@citrix.com> [xen] Acked-by: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com> [xen swiotlb] Acked-by: Joerg Roedel <jroedel@suse.de> [iommu] Acked-by: Richard Kuo <rkuo@codeaurora.org> [hexagon] Acked-by: Geert Uytterhoeven <geert@linux-m68k.org> [m68k] Acked-by: Gerald Schaefer <gerald.schaefer@de.ibm.com> [s390] Acked-by: Bjorn Andersson <bjorn.andersson@linaro.org> Acked-by: Hans-Christian Noren Egtvedt <egtvedt@samfundet.no> [avr32] Acked-by: Vineet Gupta <vgupta@synopsys.com> [arc] Acked-by: Robin Murphy <robin.murphy@arm.com> [arm64 and dma-iommu] Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-08-03 14:46:00 -06:00
attrs);
dev_err(fbdev->dev, "vrfb create ctx failed\n");
return r;
}
}
rg->attrs = attrs;
rg->token = token;
rg->dma_handle = dma_handle;
rg->paddr = (unsigned long)dma_handle;
rg->vaddr = (void __iomem *)token;
rg->size = size;
rg->alloc = 1;
return 0;
}
/* allocate fbmem using display resolution as reference */
static int omapfb_alloc_fbmem_display(struct fb_info *fbi, unsigned long size,
unsigned long paddr)
{
struct omapfb_info *ofbi = FB2OFB(fbi);
struct omapfb2_device *fbdev = ofbi->fbdev;
struct omap_dss_device *display;
int bytespp;
display = fb2display(fbi);
if (!display)
return 0;
switch (omapfb_get_recommended_bpp(fbdev, display)) {
case 16:
bytespp = 2;
break;
case 24:
bytespp = 4;
break;
default:
bytespp = 4;
break;
}
if (!size) {
u16 w, h;
display->driver->get_resolution(display, &w, &h);
if (ofbi->rotation_type == OMAP_DSS_ROT_VRFB) {
size = max(omap_vrfb_min_phys_size(w, h, bytespp),
omap_vrfb_min_phys_size(h, w, bytespp));
DBG("adjusting fb mem size for VRFB, %u -> %lu\n",
w * h * bytespp, size);
} else {
size = w * h * bytespp;
}
}
if (!size)
return 0;
return omapfb_alloc_fbmem(fbi, size, paddr);
}
static int omapfb_parse_vram_param(const char *param, int max_entries,
unsigned long *sizes, unsigned long *paddrs)
{
unsigned int fbnum;
unsigned long size;
unsigned long paddr = 0;
char *p, *start;
start = (char *)param;
while (1) {
p = start;
fbnum = simple_strtoul(p, &p, 10);
if (p == start)
return -EINVAL;
if (*p != ':')
return -EINVAL;
if (fbnum >= max_entries)
return -EINVAL;
size = memparse(p + 1, &p);
if (!size)
return -EINVAL;
paddr = 0;
if (*p == '@') {
paddr = simple_strtoul(p + 1, &p, 16);
if (!paddr)
return -EINVAL;
}
WARN_ONCE(paddr,
"reserving memory at predefined address not supported\n");
paddrs[fbnum] = paddr;
sizes[fbnum] = size;
if (*p == 0)
break;
if (*p != ',')
return -EINVAL;
++p;
start = p;
}
return 0;
}
static int omapfb_allocate_all_fbs(struct omapfb2_device *fbdev)
{
int i, r;
unsigned long vram_sizes[10];
unsigned long vram_paddrs[10];
memset(&vram_sizes, 0, sizeof(vram_sizes));
memset(&vram_paddrs, 0, sizeof(vram_paddrs));
if (def_vram && omapfb_parse_vram_param(def_vram, 10,
vram_sizes, vram_paddrs)) {
dev_err(fbdev->dev, "failed to parse vram parameter\n");
memset(&vram_sizes, 0, sizeof(vram_sizes));
memset(&vram_paddrs, 0, sizeof(vram_paddrs));
}
for (i = 0; i < fbdev->num_fbs; i++) {
/* allocate memory automatically only for fb0, or if
* excplicitly defined with vram or plat data option */
if (i == 0 || vram_sizes[i] != 0) {
r = omapfb_alloc_fbmem_display(fbdev->fbs[i],
vram_sizes[i], vram_paddrs[i]);
if (r)
return r;
}
}
for (i = 0; i < fbdev->num_fbs; i++) {
struct omapfb_info *ofbi = FB2OFB(fbdev->fbs[i]);
struct omapfb2_mem_region *rg;
rg = ofbi->region;
DBG("region%d phys %08x virt %p size=%lu\n",
i,
rg->paddr,
rg->vaddr,
rg->size);
}
return 0;
}
static void omapfb_clear_fb(struct fb_info *fbi)
{
const struct fb_fillrect rect = {
.dx = 0,
.dy = 0,
.width = fbi->var.xres_virtual,
.height = fbi->var.yres_virtual,
.color = 0,
.rop = ROP_COPY,
};
cfb_fillrect(fbi, &rect);
}
int omapfb_realloc_fbmem(struct fb_info *fbi, unsigned long size, int type)
{
struct omapfb_info *ofbi = FB2OFB(fbi);
struct omapfb2_device *fbdev = ofbi->fbdev;
struct omapfb2_mem_region *rg = ofbi->region;
unsigned long old_size = rg->size;
unsigned long old_paddr = rg->paddr;
int old_type = rg->type;
int r;
if (type != OMAPFB_MEMTYPE_SDRAM)
return -EINVAL;
size = PAGE_ALIGN(size);
if (old_size == size && old_type == type)
return 0;
omapfb_free_fbmem(fbi);
if (size == 0) {
clear_fb_info(fbi);
return 0;
}
r = omapfb_alloc_fbmem(fbi, size, 0);
if (r) {
if (old_size)
omapfb_alloc_fbmem(fbi, old_size, old_paddr);
if (rg->size == 0)
clear_fb_info(fbi);
return r;
}
if (old_size == size)
return 0;
if (old_size == 0) {
DBG("initializing fb %d\n", ofbi->id);
r = omapfb_fb_init(fbdev, fbi);
if (r) {
DBG("omapfb_fb_init failed\n");
goto err;
}
r = omapfb_apply_changes(fbi, 1);
if (r) {
DBG("omapfb_apply_changes failed\n");
goto err;
}
} else {
struct fb_var_screeninfo new_var;
memcpy(&new_var, &fbi->var, sizeof(new_var));
r = check_fb_var(fbi, &new_var);
if (r)
goto err;
memcpy(&fbi->var, &new_var, sizeof(fbi->var));
set_fb_fix(fbi);
r = setup_vrfb_rotation(fbi);
if (r)
goto err;
}
omapfb_clear_fb(fbi);
return 0;
err:
omapfb_free_fbmem(fbi);
clear_fb_info(fbi);
return r;
}
static void omapfb_auto_update_work(struct work_struct *work)
{
struct omap_dss_device *dssdev;
struct omap_dss_driver *dssdrv;
struct omapfb_display_data *d;
u16 w, h;
unsigned int freq;
struct omapfb2_device *fbdev;
d = container_of(work, struct omapfb_display_data,
auto_update_work.work);
dssdev = d->dssdev;
dssdrv = dssdev->driver;
fbdev = d->fbdev;
if (!dssdrv || !dssdrv->update)
return;
if (dssdrv->sync)
dssdrv->sync(dssdev);
dssdrv->get_resolution(dssdev, &w, &h);
dssdrv->update(dssdev, 0, 0, w, h);
freq = auto_update_freq;
if (freq == 0)
freq = 20;
queue_delayed_work(fbdev->auto_update_wq,
&d->auto_update_work, HZ / freq);
}
void omapfb_start_auto_update(struct omapfb2_device *fbdev,
struct omap_dss_device *display)
{
struct omapfb_display_data *d;
if (fbdev->auto_update_wq == NULL) {
struct workqueue_struct *wq;
wq = create_singlethread_workqueue("omapfb_auto_update");
if (wq == NULL) {
dev_err(fbdev->dev, "Failed to create workqueue for "
"auto-update\n");
return;
}
fbdev->auto_update_wq = wq;
}
d = get_display_data(fbdev, display);
INIT_DELAYED_WORK(&d->auto_update_work, omapfb_auto_update_work);
d->auto_update_work_enabled = true;
omapfb_auto_update_work(&d->auto_update_work.work);
}
void omapfb_stop_auto_update(struct omapfb2_device *fbdev,
struct omap_dss_device *display)
{
struct omapfb_display_data *d;
d = get_display_data(fbdev, display);
cancel_delayed_work_sync(&d->auto_update_work);
d->auto_update_work_enabled = false;
}
/* initialize fb_info, var, fix to something sane based on the display */
static int omapfb_fb_init(struct omapfb2_device *fbdev, struct fb_info *fbi)
{
struct fb_var_screeninfo *var = &fbi->var;
struct omap_dss_device *display = fb2display(fbi);
struct omapfb_info *ofbi = FB2OFB(fbi);
int r = 0;
fbi->fbops = &omapfb_ops;
fbi->flags = FBINFO_FLAG_DEFAULT;
fbi->pseudo_palette = fbdev->pseudo_palette;
if (ofbi->region->size == 0) {
clear_fb_info(fbi);
return 0;
}
var->nonstd = 0;
var->bits_per_pixel = 0;
var->rotate = def_rotate;
if (display) {
u16 w, h;
int rotation = (var->rotate + ofbi->rotation[0]) % 4;
display->driver->get_resolution(display, &w, &h);
if (rotation == FB_ROTATE_CW ||
rotation == FB_ROTATE_CCW) {
var->xres = h;
var->yres = w;
} else {
var->xres = w;
var->yres = h;
}
var->xres_virtual = var->xres;
var->yres_virtual = var->yres;
if (!var->bits_per_pixel) {
switch (omapfb_get_recommended_bpp(fbdev, display)) {
case 16:
var->bits_per_pixel = 16;
break;
case 24:
var->bits_per_pixel = 32;
break;
default:
dev_err(fbdev->dev, "illegal display "
"bpp\n");
return -EINVAL;
}
}
} else {
/* if there's no display, let's just guess some basic values */
var->xres = 320;
var->yres = 240;
var->xres_virtual = var->xres;
var->yres_virtual = var->yres;
if (!var->bits_per_pixel)
var->bits_per_pixel = 16;
}
r = check_fb_var(fbi, var);
if (r)
goto err;
set_fb_fix(fbi);
r = setup_vrfb_rotation(fbi);
if (r)
goto err;
r = fb_alloc_cmap(&fbi->cmap, 256, 0);
if (r)
dev_err(fbdev->dev, "unable to allocate color map memory\n");
err:
return r;
}
static void fbinfo_cleanup(struct omapfb2_device *fbdev, struct fb_info *fbi)
{
fb_dealloc_cmap(&fbi->cmap);
}
static void omapfb_free_resources(struct omapfb2_device *fbdev)
{
int i;
DBG("free_resources\n");
if (fbdev == NULL)
return;
for (i = 0; i < fbdev->num_overlays; i++) {
struct omap_overlay *ovl = fbdev->overlays[i];
ovl->disable(ovl);
if (ovl->manager)
ovl->unset_manager(ovl);
}
for (i = 0; i < fbdev->num_fbs; i++)
unregister_framebuffer(fbdev->fbs[i]);
/* free the reserved fbmem */
omapfb_free_all_fbmem(fbdev);
for (i = 0; i < fbdev->num_fbs; i++) {
fbinfo_cleanup(fbdev, fbdev->fbs[i]);
framebuffer_release(fbdev->fbs[i]);
}
for (i = 0; i < fbdev->num_displays; i++) {
struct omap_dss_device *dssdev = fbdev->displays[i].dssdev;
if (fbdev->displays[i].auto_update_work_enabled)
omapfb_stop_auto_update(fbdev, dssdev);
if (dssdev->state != OMAP_DSS_DISPLAY_DISABLED)
dssdev->driver->disable(dssdev);
OMAPDSS: Implement display (dis)connect support We currently have two steps in panel initialization and startup: probing and enabling. After the panel has been probed, it's ready and can be configured and later enabled. This model is not enough with more complex display pipelines, where we may have, for example, two panels, of which only one can be used at a time, connected to the same video output. To support that kind of scenarios, we need to add new step to the initialization: connect. This patch adds support for connecting and disconnecting panels. After probe, but before connect, no panel ops should be called. When the connect is called, a proper video pipeline is established, and the panel is ready for use. If some part in the video pipeline is already connected (by some other panel), the connect call fails. One key difference with the old style setup is that connect() handles also connecting to the overlay manager. This means that the omapfb (or omapdrm) no longer needs to figure out which overlay manager to use, but it can just call connect() on the panel, and the proper overlay manager is connected by omapdss. This also allows us to add back the support for dynamic switching between two exclusive panels. However, the current panel device model is not changed to support this, as the new device model is implemented in the following patches and the old model will be removed. The new device model supports dynamic switching. Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ti.com>
2013-05-08 07:23:32 -06:00
dssdev->driver->disconnect(dssdev);
omap_dss_put_device(dssdev);
}
if (fbdev->auto_update_wq != NULL) {
flush_workqueue(fbdev->auto_update_wq);
destroy_workqueue(fbdev->auto_update_wq);
fbdev->auto_update_wq = NULL;
}
dev_set_drvdata(fbdev->dev, NULL);
}
static int omapfb_create_framebuffers(struct omapfb2_device *fbdev)
{
int r, i;
fbdev->num_fbs = 0;
DBG("create %d framebuffers\n", CONFIG_FB_OMAP2_NUM_FBS);
/* allocate fb_infos */
for (i = 0; i < CONFIG_FB_OMAP2_NUM_FBS; i++) {
struct fb_info *fbi;
struct omapfb_info *ofbi;
fbi = framebuffer_alloc(sizeof(struct omapfb_info),
fbdev->dev);
if (fbi == NULL) {
dev_err(fbdev->dev,
"unable to allocate memory for plane info\n");
return -ENOMEM;
}
clear_fb_info(fbi);
fbdev->fbs[i] = fbi;
ofbi = FB2OFB(fbi);
ofbi->fbdev = fbdev;
ofbi->id = i;
ofbi->region = &fbdev->regions[i];
ofbi->region->id = i;
init_rwsem(&ofbi->region->lock);
/* assign these early, so that fb alloc can use them */
ofbi->rotation_type = def_vrfb ? OMAP_DSS_ROT_VRFB :
OMAP_DSS_ROT_DMA;
ofbi->mirror = def_mirror;
fbdev->num_fbs++;
}
DBG("fb_infos allocated\n");
/* assign overlays for the fbs */
for (i = 0; i < min(fbdev->num_fbs, fbdev->num_overlays); i++) {
struct omapfb_info *ofbi = FB2OFB(fbdev->fbs[i]);
ofbi->overlays[0] = fbdev->overlays[i];
ofbi->num_overlays = 1;
}
/* allocate fb memories */
r = omapfb_allocate_all_fbs(fbdev);
if (r) {
dev_err(fbdev->dev, "failed to allocate fbmem\n");
return r;
}
DBG("fbmems allocated\n");
/* setup fb_infos */
for (i = 0; i < fbdev->num_fbs; i++) {
struct fb_info *fbi = fbdev->fbs[i];
struct omapfb_info *ofbi = FB2OFB(fbi);
omapfb_get_mem_region(ofbi->region);
r = omapfb_fb_init(fbdev, fbi);
omapfb_put_mem_region(ofbi->region);
if (r) {
dev_err(fbdev->dev, "failed to setup fb_info\n");
return r;
}
}
for (i = 0; i < fbdev->num_fbs; i++) {
struct fb_info *fbi = fbdev->fbs[i];
struct omapfb_info *ofbi = FB2OFB(fbi);
if (ofbi->region->size == 0)
continue;
omapfb_clear_fb(fbi);
}
DBG("fb_infos initialized\n");
for (i = 0; i < fbdev->num_fbs; i++) {
r = register_framebuffer(fbdev->fbs[i]);
if (r != 0) {
dev_err(fbdev->dev,
"registering framebuffer %d failed\n", i);
return r;
}
}
DBG("framebuffers registered\n");
for (i = 0; i < fbdev->num_fbs; i++) {
struct fb_info *fbi = fbdev->fbs[i];
struct omapfb_info *ofbi = FB2OFB(fbi);
omapfb_get_mem_region(ofbi->region);
r = omapfb_apply_changes(fbi, 1);
omapfb_put_mem_region(ofbi->region);
if (r) {
dev_err(fbdev->dev, "failed to change mode\n");
return r;
}
}
/* Enable fb0 */
if (fbdev->num_fbs > 0) {
struct omapfb_info *ofbi = FB2OFB(fbdev->fbs[0]);
if (ofbi->num_overlays > 0) {
struct omap_overlay *ovl = ofbi->overlays[0];
ovl->manager->apply(ovl->manager);
r = omapfb_overlay_enable(ovl, 1);
if (r) {
dev_err(fbdev->dev,
"failed to enable overlay\n");
return r;
}
}
}
DBG("create_framebuffers done\n");
return 0;
}
static int omapfb_mode_to_timings(const char *mode_str,
struct omap_dss_device *display,
struct omap_video_timings *timings, u8 *bpp)
{
struct fb_info *fbi;
struct fb_var_screeninfo *var;
struct fb_ops *fbops;
int r;
#ifdef CONFIG_OMAP2_DSS_VENC
if (strcmp(mode_str, "pal") == 0) {
*timings = omap_dss_pal_timings;
*bpp = 24;
return 0;
} else if (strcmp(mode_str, "ntsc") == 0) {
*timings = omap_dss_ntsc_timings;
*bpp = 24;
return 0;
}
#endif
/* this is quite a hack, but I wanted to use the modedb and for
* that we need fb_info and var, so we create dummy ones */
*bpp = 0;
fbi = NULL;
var = NULL;
fbops = NULL;
fbi = kzalloc(sizeof(*fbi), GFP_KERNEL);
if (fbi == NULL) {
r = -ENOMEM;
goto err;
}
var = kzalloc(sizeof(*var), GFP_KERNEL);
if (var == NULL) {
r = -ENOMEM;
goto err;
}
fbops = kzalloc(sizeof(*fbops), GFP_KERNEL);
if (fbops == NULL) {
r = -ENOMEM;
goto err;
}
fbi->fbops = fbops;
r = fb_find_mode(var, fbi, mode_str, NULL, 0, NULL, 24);
if (r == 0) {
r = -EINVAL;
goto err;
}
if (display->driver->get_timings) {
display->driver->get_timings(display, timings);
} else {
timings->data_pclk_edge = OMAPDSS_DRIVE_SIG_RISING_EDGE;
timings->de_level = OMAPDSS_SIG_ACTIVE_HIGH;
timings->sync_pclk_edge = OMAPDSS_DRIVE_SIG_FALLING_EDGE;
}
timings->pixelclock = PICOS2KHZ(var->pixclock) * 1000;
timings->hbp = var->left_margin;
timings->hfp = var->right_margin;
timings->vbp = var->upper_margin;
timings->vfp = var->lower_margin;
timings->hsw = var->hsync_len;
timings->vsw = var->vsync_len;
timings->x_res = var->xres;
timings->y_res = var->yres;
timings->hsync_level = var->sync & FB_SYNC_HOR_HIGH_ACT ?
OMAPDSS_SIG_ACTIVE_HIGH :
OMAPDSS_SIG_ACTIVE_LOW;
timings->vsync_level = var->sync & FB_SYNC_VERT_HIGH_ACT ?
OMAPDSS_SIG_ACTIVE_HIGH :
OMAPDSS_SIG_ACTIVE_LOW;
timings->interlace = var->vmode & FB_VMODE_INTERLACED;
switch (var->bits_per_pixel) {
case 16:
*bpp = 16;
break;
case 24:
case 32:
default:
*bpp = 24;
break;
}
r = 0;
err:
kfree(fbi);
kfree(var);
kfree(fbops);
return r;
}
static int omapfb_set_def_mode(struct omapfb2_device *fbdev,
struct omap_dss_device *display, char *mode_str)
{
int r;
u8 bpp;
struct omap_video_timings timings, temp_timings;
struct omapfb_display_data *d;
r = omapfb_mode_to_timings(mode_str, display, &timings, &bpp);
if (r)
return r;
d = get_display_data(fbdev, display);
d->bpp_override = bpp;
if (display->driver->check_timings) {
r = display->driver->check_timings(display, &timings);
if (r)
return r;
} else {
/* If check_timings is not present compare xres and yres */
if (display->driver->get_timings) {
display->driver->get_timings(display, &temp_timings);
if (temp_timings.x_res != timings.x_res ||
temp_timings.y_res != timings.y_res)
return -EINVAL;
}
}
if (display->driver->set_timings)
display->driver->set_timings(display, &timings);
return 0;
}
static int omapfb_get_recommended_bpp(struct omapfb2_device *fbdev,
struct omap_dss_device *dssdev)
{
struct omapfb_display_data *d;
BUG_ON(dssdev->driver->get_recommended_bpp == NULL);
d = get_display_data(fbdev, dssdev);
if (d->bpp_override != 0)
return d->bpp_override;
return dssdev->driver->get_recommended_bpp(dssdev);
}
static int omapfb_parse_def_modes(struct omapfb2_device *fbdev)
{
char *str, *options, *this_opt;
int r = 0;
str = kstrdup(def_mode, GFP_KERNEL);
if (!str)
return -ENOMEM;
options = str;
while (!r && (this_opt = strsep(&options, ",")) != NULL) {
char *p, *display_str, *mode_str;
struct omap_dss_device *display;
int i;
p = strchr(this_opt, ':');
if (!p) {
r = -EINVAL;
break;
}
*p = 0;
display_str = this_opt;
mode_str = p + 1;
display = NULL;
for (i = 0; i < fbdev->num_displays; ++i) {
if (strcmp(fbdev->displays[i].dssdev->name,
display_str) == 0) {
display = fbdev->displays[i].dssdev;
break;
}
}
if (!display) {
r = -EINVAL;
break;
}
r = omapfb_set_def_mode(fbdev, display, mode_str);
if (r)
break;
}
kfree(str);
return r;
}
static void fb_videomode_to_omap_timings(struct fb_videomode *m,
struct omap_dss_device *display,
struct omap_video_timings *t)
{
if (display->driver->get_timings) {
display->driver->get_timings(display, t);
} else {
t->data_pclk_edge = OMAPDSS_DRIVE_SIG_RISING_EDGE;
t->de_level = OMAPDSS_SIG_ACTIVE_HIGH;
t->sync_pclk_edge = OMAPDSS_DRIVE_SIG_FALLING_EDGE;
}
t->x_res = m->xres;
t->y_res = m->yres;
t->pixelclock = PICOS2KHZ(m->pixclock) * 1000;
t->hsw = m->hsync_len;
t->hfp = m->right_margin;
t->hbp = m->left_margin;
t->vsw = m->vsync_len;
t->vfp = m->lower_margin;
t->vbp = m->upper_margin;
t->hsync_level = m->sync & FB_SYNC_HOR_HIGH_ACT ?
OMAPDSS_SIG_ACTIVE_HIGH :
OMAPDSS_SIG_ACTIVE_LOW;
t->vsync_level = m->sync & FB_SYNC_VERT_HIGH_ACT ?
OMAPDSS_SIG_ACTIVE_HIGH :
OMAPDSS_SIG_ACTIVE_LOW;
t->interlace = m->vmode & FB_VMODE_INTERLACED;
}
static int omapfb_find_best_mode(struct omap_dss_device *display,
struct omap_video_timings *timings)
{
struct fb_monspecs *specs;
u8 *edid;
int r, i, best_idx, len;
if (!display->driver->read_edid)
return -ENODEV;
len = 0x80 * 2;
edid = kmalloc(len, GFP_KERNEL);
if (edid == NULL)
return -ENOMEM;
r = display->driver->read_edid(display, edid, len);
if (r < 0)
goto err1;
specs = kzalloc(sizeof(*specs), GFP_KERNEL);
if (specs == NULL) {
r = -ENOMEM;
goto err1;
}
fb_edid_to_monspecs(edid, specs);
best_idx = -1;
for (i = 0; i < specs->modedb_len; ++i) {
struct fb_videomode *m;
struct omap_video_timings t;
m = &specs->modedb[i];
if (m->pixclock == 0)
continue;
/* skip repeated pixel modes */
if (m->xres == 2880 || m->xres == 1440)
continue;
if (m->vmode & FB_VMODE_INTERLACED ||
m->vmode & FB_VMODE_DOUBLE)
continue;
fb_videomode_to_omap_timings(m, display, &t);
r = display->driver->check_timings(display, &t);
if (r == 0) {
best_idx = i;
break;
}
}
if (best_idx == -1) {
r = -ENOENT;
goto err2;
}
fb_videomode_to_omap_timings(&specs->modedb[best_idx], display,
timings);
r = 0;
err2:
fb_destroy_modedb(specs->modedb);
kfree(specs);
err1:
kfree(edid);
return r;
}
static int omapfb_init_display(struct omapfb2_device *fbdev,
struct omap_dss_device *dssdev)
{
struct omap_dss_driver *dssdrv = dssdev->driver;
struct omapfb_display_data *d;
int r;
r = dssdrv->enable(dssdev);
if (r) {
dev_warn(fbdev->dev, "Failed to enable display '%s'\n",
dssdev->name);
return r;
}
d = get_display_data(fbdev, dssdev);
d->fbdev = fbdev;
if (dssdev->caps & OMAP_DSS_DISPLAY_CAP_MANUAL_UPDATE) {
u16 w, h;
if (auto_update) {
omapfb_start_auto_update(fbdev, dssdev);
d->update_mode = OMAPFB_AUTO_UPDATE;
} else {
d->update_mode = OMAPFB_MANUAL_UPDATE;
}
if (dssdrv->enable_te) {
r = dssdrv->enable_te(dssdev, 1);
if (r) {
dev_err(fbdev->dev, "Failed to set TE\n");
return r;
}
}
dssdrv->get_resolution(dssdev, &w, &h);
r = dssdrv->update(dssdev, 0, 0, w, h);
if (r) {
dev_err(fbdev->dev,
"Failed to update display\n");
return r;
}
} else {
d->update_mode = OMAPFB_AUTO_UPDATE;
}
return 0;
}
static int omapfb_init_connections(struct omapfb2_device *fbdev,
struct omap_dss_device *def_dssdev)
{
int i, r;
struct omap_overlay_manager *mgr;
OMAPDSS: Implement display (dis)connect support We currently have two steps in panel initialization and startup: probing and enabling. After the panel has been probed, it's ready and can be configured and later enabled. This model is not enough with more complex display pipelines, where we may have, for example, two panels, of which only one can be used at a time, connected to the same video output. To support that kind of scenarios, we need to add new step to the initialization: connect. This patch adds support for connecting and disconnecting panels. After probe, but before connect, no panel ops should be called. When the connect is called, a proper video pipeline is established, and the panel is ready for use. If some part in the video pipeline is already connected (by some other panel), the connect call fails. One key difference with the old style setup is that connect() handles also connecting to the overlay manager. This means that the omapfb (or omapdrm) no longer needs to figure out which overlay manager to use, but it can just call connect() on the panel, and the proper overlay manager is connected by omapdss. This also allows us to add back the support for dynamic switching between two exclusive panels. However, the current panel device model is not changed to support this, as the new device model is implemented in the following patches and the old model will be removed. The new device model supports dynamic switching. Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ti.com>
2013-05-08 07:23:32 -06:00
r = def_dssdev->driver->connect(def_dssdev);
if (r) {
dev_err(fbdev->dev, "failed to connect default display\n");
return r;
}
for (i = 0; i < fbdev->num_displays; ++i) {
struct omap_dss_device *dssdev = fbdev->displays[i].dssdev;
OMAPDSS: Implement display (dis)connect support We currently have two steps in panel initialization and startup: probing and enabling. After the panel has been probed, it's ready and can be configured and later enabled. This model is not enough with more complex display pipelines, where we may have, for example, two panels, of which only one can be used at a time, connected to the same video output. To support that kind of scenarios, we need to add new step to the initialization: connect. This patch adds support for connecting and disconnecting panels. After probe, but before connect, no panel ops should be called. When the connect is called, a proper video pipeline is established, and the panel is ready for use. If some part in the video pipeline is already connected (by some other panel), the connect call fails. One key difference with the old style setup is that connect() handles also connecting to the overlay manager. This means that the omapfb (or omapdrm) no longer needs to figure out which overlay manager to use, but it can just call connect() on the panel, and the proper overlay manager is connected by omapdss. This also allows us to add back the support for dynamic switching between two exclusive panels. However, the current panel device model is not changed to support this, as the new device model is implemented in the following patches and the old model will be removed. The new device model supports dynamic switching. Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ti.com>
2013-05-08 07:23:32 -06:00
if (dssdev == def_dssdev)
continue;
OMAPDSS: Implement display (dis)connect support We currently have two steps in panel initialization and startup: probing and enabling. After the panel has been probed, it's ready and can be configured and later enabled. This model is not enough with more complex display pipelines, where we may have, for example, two panels, of which only one can be used at a time, connected to the same video output. To support that kind of scenarios, we need to add new step to the initialization: connect. This patch adds support for connecting and disconnecting panels. After probe, but before connect, no panel ops should be called. When the connect is called, a proper video pipeline is established, and the panel is ready for use. If some part in the video pipeline is already connected (by some other panel), the connect call fails. One key difference with the old style setup is that connect() handles also connecting to the overlay manager. This means that the omapfb (or omapdrm) no longer needs to figure out which overlay manager to use, but it can just call connect() on the panel, and the proper overlay manager is connected by omapdss. This also allows us to add back the support for dynamic switching between two exclusive panels. However, the current panel device model is not changed to support this, as the new device model is implemented in the following patches and the old model will be removed. The new device model supports dynamic switching. Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ti.com>
2013-05-08 07:23:32 -06:00
/*
* We don't care if the connect succeeds or not. We just want to
* connect as many displays as possible.
*/
dssdev->driver->connect(dssdev);
}
mgr = omapdss_find_mgr_from_display(def_dssdev);
if (!mgr) {
dev_err(fbdev->dev, "no ovl manager for the default display\n");
return -EINVAL;
}
for (i = 0; i < fbdev->num_overlays; i++) {
struct omap_overlay *ovl = fbdev->overlays[i];
if (ovl->manager)
ovl->unset_manager(ovl);
r = ovl->set_manager(ovl, mgr);
if (r)
dev_warn(fbdev->dev,
"failed to connect overlay %s to manager %s\n",
ovl->name, mgr->name);
}
return 0;
}
static struct omap_dss_device *
omapfb_find_default_display(struct omapfb2_device *fbdev)
{
const char *def_name;
int i;
/*
* Search with the display name from the user or the board file,
* comparing to display names and aliases
*/
def_name = omapdss_get_default_display_name();
if (def_name) {
for (i = 0; i < fbdev->num_displays; ++i) {
struct omap_dss_device *dssdev;
dssdev = fbdev->displays[i].dssdev;
if (dssdev->name && strcmp(def_name, dssdev->name) == 0)
return dssdev;
if (strcmp(def_name, dssdev->alias) == 0)
return dssdev;
}
/* def_name given but not found */
return NULL;
}
/* then look for DT alias display0 */
for (i = 0; i < fbdev->num_displays; ++i) {
struct omap_dss_device *dssdev;
int id;
dssdev = fbdev->displays[i].dssdev;
if (dssdev->dev->of_node == NULL)
continue;
id = of_alias_get_id(dssdev->dev->of_node, "display");
if (id == 0)
return dssdev;
}
/* return the first display we have in the list */
return fbdev->displays[0].dssdev;
}
static int omapfb_probe(struct platform_device *pdev)
{
struct omapfb2_device *fbdev = NULL;
int r = 0;
int i;
struct omap_dss_device *def_display;
struct omap_dss_device *dssdev;
DBG("omapfb_probe\n");
if (omapdss_is_initialized() == false)
return -EPROBE_DEFER;
if (pdev->num_resources != 0) {
dev_err(&pdev->dev, "probed for an unknown device\n");
r = -ENODEV;
goto err0;
}
fbdev = devm_kzalloc(&pdev->dev, sizeof(struct omapfb2_device),
GFP_KERNEL);
if (fbdev == NULL) {
r = -ENOMEM;
goto err0;
}
if (def_vrfb && !omap_vrfb_supported()) {
def_vrfb = 0;
dev_warn(&pdev->dev, "VRFB is not supported on this hardware, "
"ignoring the module parameter vrfb=y\n");
}
r = omapdss_compat_init();
if (r)
goto err0;
mutex_init(&fbdev->mtx);
fbdev->dev = &pdev->dev;
platform_set_drvdata(pdev, fbdev);
fbdev->num_displays = 0;
dssdev = NULL;
for_each_dss_dev(dssdev) {
struct omapfb_display_data *d;
omap_dss_get_device(dssdev);
if (!dssdev->driver) {
dev_warn(&pdev->dev, "no driver for display: %s\n",
dssdev->name);
omap_dss_put_device(dssdev);
continue;
}
d = &fbdev->displays[fbdev->num_displays++];
d->dssdev = dssdev;
if (dssdev->caps & OMAP_DSS_DISPLAY_CAP_MANUAL_UPDATE)
d->update_mode = OMAPFB_MANUAL_UPDATE;
else
d->update_mode = OMAPFB_AUTO_UPDATE;
}
if (fbdev->num_displays == 0) {
dev_err(&pdev->dev, "no displays\n");
r = -EPROBE_DEFER;
goto cleanup;
}
fbdev->num_overlays = omap_dss_get_num_overlays();
for (i = 0; i < fbdev->num_overlays; i++)
fbdev->overlays[i] = omap_dss_get_overlay(i);
fbdev->num_managers = omap_dss_get_num_overlay_managers();
for (i = 0; i < fbdev->num_managers; i++)
fbdev->managers[i] = omap_dss_get_overlay_manager(i);
def_display = omapfb_find_default_display(fbdev);
if (def_display == NULL) {
dev_err(fbdev->dev, "failed to find default display\n");
r = -EPROBE_DEFER;
goto cleanup;
}
r = omapfb_init_connections(fbdev, def_display);
if (r) {
dev_err(fbdev->dev, "failed to init overlay connections\n");
goto cleanup;
}
if (def_mode && strlen(def_mode) > 0) {
if (omapfb_parse_def_modes(fbdev))
dev_warn(&pdev->dev, "cannot parse default modes\n");
} else if (def_display && def_display->driver->set_timings &&
def_display->driver->check_timings) {
struct omap_video_timings t;
r = omapfb_find_best_mode(def_display, &t);
if (r == 0)
def_display->driver->set_timings(def_display, &t);
}
r = omapfb_create_framebuffers(fbdev);
if (r)
goto cleanup;
for (i = 0; i < fbdev->num_managers; i++) {
struct omap_overlay_manager *mgr;
mgr = fbdev->managers[i];
r = mgr->apply(mgr);
if (r)
dev_warn(fbdev->dev, "failed to apply dispc config\n");
}
DBG("mgr->apply'ed\n");
if (def_display) {
r = omapfb_init_display(fbdev, def_display);
if (r) {
dev_err(fbdev->dev,
"failed to initialize default "
"display\n");
goto cleanup;
}
}
DBG("create sysfs for fbs\n");
r = omapfb_create_sysfs(fbdev);
if (r) {
dev_err(fbdev->dev, "failed to create sysfs entries\n");
goto cleanup;
}
if (def_display) {
u16 w, h;
def_display->driver->get_resolution(def_display, &w, &h);
dev_info(fbdev->dev, "using display '%s' mode %dx%d\n",
def_display->name, w, h);
}
return 0;
cleanup:
omapfb_free_resources(fbdev);
omapdss_compat_uninit();
err0:
dev_err(&pdev->dev, "failed to setup omapfb\n");
return r;
}
static int omapfb_remove(struct platform_device *pdev)
{
struct omapfb2_device *fbdev = platform_get_drvdata(pdev);
/* FIXME: wait till completion of pending events */
omapfb_remove_sysfs(fbdev);
omapfb_free_resources(fbdev);
omapdss_compat_uninit();
return 0;
}
static struct platform_driver omapfb_driver = {
.probe = omapfb_probe,
.remove = omapfb_remove,
.driver = {
.name = "omapfb",
},
};
module_param_named(mode, def_mode, charp, 0);
module_param_named(vram, def_vram, charp, 0);
module_param_named(rotate, def_rotate, int, 0);
module_param_named(vrfb, def_vrfb, bool, 0);
module_param_named(mirror, def_mirror, bool, 0);
module_platform_driver(omapfb_driver);
MODULE_ALIAS("platform:omapfb");
MODULE_AUTHOR("Tomi Valkeinen <tomi.valkeinen@nokia.com>");
MODULE_DESCRIPTION("OMAP2/3 Framebuffer");
MODULE_LICENSE("GPL v2");